The HP had twice as many USB ports and two of them are version 3.0 (vs. MacPro's 2.0 only support.) Yes, there is probably some processor performance differences, but I would guess not that much.
*Gasp* Twice as many USB ports?
If Apple added 10 USB ports to the front, didn't update the GPU or CPU at all, would you consider that a successful refresh?
I'm starting to think that Apple's view of a pro user differs from what most people here using a Mac Pro would view as a pro user. I'm sorry to say that I just don't see any of the so called Macbook Pro machines as a Pro machine. "Pro" is just a product differentiator now that commands a higher cost, it no longer signifies the type of user. Any machine (Retina MBP currently, the others will follow) that locks you in a set quantity of RAM and storage at purchase time is just a large iPad - it is a consumer device and nothing more. And that is the direction I think they may be taking with the Mac Pro, the exciting new 2013 Mac Pro could effectively end up being a slightly bigger Mac Mini, maybe with a desktop graphics card but with far less expansion potential than the current machine.
I don't necessarily think this is true - first, the storage isn't necessarily locked in (I have faith the OWC guys will have replacement SSDs for the rMBP). I think calling a rMBP a "large iPad" is extremely disingenuous - it's a very, very powerful machine, and absolutely suited for a large number of "Pro" applications in a way either a more consumer oriented laptop (the Air) or an iPad isn't even close.
I could comfortably run the entirety of my research from a rMBP, with the exception of the stuff that belongs on a cluster. That, to me, defines a Pro laptop - being able to pop more RAM in is nice, but not necessary.
I think what the "Apple could design it!" people are missing is that because Apple can do something doesn't mean it is profitable or wise to do so. Right now, they let Intel (rightly) do most of the heavy lifting for their Mac Pro R&D, using their designs. While Apple could bring that in house, that involves ramping up an entire design branch that Apple doesn't necessarily have. That's incredibly costly when you compare it to "Call us when you have something Intel", and for a niche product, may not actually be worthwhile.
Basically, the argument for the survival of the Pro as a niche product ("It's basically free to develop for Apple") is mutually incompatible with Apple freeing itself from Intel's development cycle.
What I think is actually rather awkward is that the Mac Pro market isn't professional enough. Thunderbolt isn't so much an important enterprise product at the moment, so integrating it into what is essentially a workstation/server line has got to be a fairly low priority for Intel, but a high one for Apple. That makes for awkward timing.